The rationale for vaccination with pneumococcal capsular polysaccharide (PPS) vaccines is to induce opsonic antibodies to PPS, which are required for protection against Streptococcus pneumoniae. However, available PPS-based vaccines are poorly immunogenic in many patients at the highest risk for pneumococcal infection. Based on the following evidence we hypothesize that reduced expression of immunoglobulin genes from the VH3 subgroup translates into an impaired anti-PPS response: i) antibodies to PPS are oligoclonal and use genes from the VH3 subgroup; ii) reduced VH3 expression has been reported in patients at risk for pneumococcal infection that generate poor responses to PPS vaccines. This application proposes to determine structure-function relationships for human antibodies to serotype 3 S. pneumoniae, a cause of invasive pneumococcal infection in adults and children. The specific aims are: 1) To determine the molecular structure of human mAbs to PPS 3 generated in transgenic mice reconstituted with human immunoglobulin loci; 2) To characterize the infection in mice and as opsonic or non-opsonic in vitro; 3) To use mAbs with molecular structure as defined in Aim 1 and functional efficacy as defined in Aim 2 to select peptide mimics of PPS 3 epitopes, and to use the peptides to determine if PPS-elicited antibodies in patients recognize protective, non-protective or disease-enhancing epitopes. Our studies will provide evidence to support or refute the hypothesis that pneumococcal vaccine failure results from an inability to produce antibodies to PPS epitopes with a certain molecular structure. This work will provide a new scientific knowledge base regarding structure-function relationships for antibodies to PPS 3 that can be used clinically to evaluate vaccine efficacy and to develop more immunogenic vaccines for patients at risk for pneumococcal infection.